Altered translation of GATA1 in Diamond-Blackfan anemia
The MIT Faculty has made this article openly available. Please share
how this access benefits you. Your story matters.
Citation
Ludwig, Leif S, Hanna T Gazda, Jennifer C Eng, Stephen W
Eichhorn, Prathapan Thiru, Roxanne Ghazvinian, Tracy I
George, et al. “Altered Translation of GATA1 in DiamondBlackfan Anemia.” Nature Medicine 20, no. 7 (June 22, 2014):
748–753.
As Published
http://dx.doi.org/10.1038/nm.3557
Publisher
Nature Publishing Group
Version
Author's final manuscript
Accessed
Thu May 26 00:58:46 EDT 2016
Citable Link
http://hdl.handle.net/1721.1/96702
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Detailed Terms
http://creativecommons.org/licenses/by-nc-sa/4.0/
NIH Public Access
Author Manuscript
Nat Med. Author manuscript; available in PMC 2015 January 01.
NIH-PA Author Manuscript
Published in final edited form as:
Nat Med. 2014 July ; 20(7): 748–753. doi:10.1038/nm.3557.
Altered translation of GATA1 in Diamond-Blackfan anemia
Leif S. Ludwig1,2,3,10,11, Hanna T. Gazda3,4,5, Jennifer C. Eng2,3, Stephen W. Eichhorn2,7,
Prathapan Thiru2, Roxanne Ghazvinian4, Tracy I. George8, Jason R. Gotlib9, Alan H.
Beggs4,5, Colin A. Sieff1,5, Harvey F. Lodish2,3,7, Eric S. Lander3,6,7, and Vijay G.
Sankaran1,2,3,5,*
1Division
of Hematology/Oncology, The Manton Center for Orphan Disease Research, Boston
Children’s Hospital and Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston,
Massachusetts 02115, USA
2Whitehead
3Broad
Institute for Biomedical Research, Cambridge, Massachusetts 02142, USA
Institute, Cambridge, Massachusetts 02142, USA
NIH-PA Author Manuscript
4Division
of Genetics and Program in Genomics, The Manton Center for Orphan Disease
Research, Boston Children’s Hospital, Boston, Massachusetts 02115, USA
5Department
of Pediatrics, Harvard Medical School, Boston, Massachusetts 02115, USA
6Department
of Systems Biology, Harvard Medical School, Boston, Massachusetts 02115, USA
7Department
of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts
02142, USA
8Departments
of Pathology, Stanford University School of Medicine, Stanford, California 94305,
USA
9Division
of Hematology, Stanford University School of Medicine, Stanford, California 94305, USA
10Institute
for Chemistry and Biochemistry, Freie Universität Berlin, Berlin 14195, Germany
11Charité-Universitätsmedizin
Berlin, Berlin 10117, Germany
NIH-PA Author Manuscript
Abstract
Ribosomal protein haploinsufficiency occurs in diverse human diseases including DiamondBlackfan anemia (DBA),1,2 congenital asplenia,3 and T-cell leukemia.4 Yet how mutations in such
ubiquitously expressed proteins result in cell-type and tissue specific defects remains a mystery.5
Here, we show that GATA1 mutations that reduce full-length protein levels of this critical
*
Correspondence should be addressed to V.G.S. (sankaran@broadinstitute.org).
Online Content Any additional Methods and Supplementary Information are available in the online version of the paper.
AUTHOR CONTRIBUTIONS
L.S.L., H.T.G., and V.G.S. conceived the project; L.S.L, H.T.G., J.C.E., S.E., R.G., A.H.B., C.A.S., and V.G.S. performed the
research; L.S.L., H.T.G., P.T., H.F.L., E.S.L., and V.G.S. analyzed data; T.I.G. and J.R.G. provided clinical assessments; L.S.L.,
H.F.L., E.S.L., and V.G.S. wrote the paper with input from all authors.
The microarray data can be found in the Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/) with the accession
number GSE41817.
The authors declare no competing financial interests.
Ludwig et al.
Page 2
NIH-PA Author Manuscript
hematopoietic transcription factor can cause DBA in rare instances. We show that ribosomal
protein haploinsufficiency, the more common cause of DBA, can similarly reduce translation of
GATA1 mRNA - a phenomenon that appears to result from this mRNA having a higher threshold
for initiation of translation. In primary hematopoietic cells from patients with RPS19 mutations, a
transcriptional signature of GATA1 target genes is globally and specifically reduced, confirming
that the activity, but not the mRNA level, of GATA1 is reduced in DBA patients with ribosomal
protein mutations. The defective hematopoiesis observed in DBA patients with ribosomal protein
haploinsufficiency can be at least partially overcome by increasing GATA1 protein levels. Our
results provide a paradigm by which selective defects in translation due to mutations in ubiquitous
ribosomal proteins can result in human disease.
NIH-PA Author Manuscript
Diamond-Blackfan anemia (DBA, MIM 105650) is characterized by a specific reduction in
the production of red blood (erythroid) cells and their precursors without defects in other
hematopoietic lineages.2,6 In more than 50% of cases, DBA is caused by heterozygous lossof-function mutations (haploinsufficiency) in one of 11 genes encoding ribosomal proteins.1
Moreover, recent studies have shown that haploinsufficiency of ribosomal proteins can
contribute to other cell-type specific diseases in humans, including congenital asplenia and
T-cell lymphocytic leukemia.3,4 It remains entirely mysterious how mutations that halve the
quantity of ubiquitously expressed ribosomal proteins result in such specific human
disorders. Numerous theories have been proposed for the pathogenesis underlying these
diseases.7 However, these models are unable to explain the exquisite cell-type specificity of
DBA and the other ribosomal disorders. The experimental evidence to support such
pathogenic models for DBA is often contradictory.8 Animal models of DBA do not
faithfully mimic the disease and the involvement of different molecular pathways in the
hematopoietic defects observed is variable.9,10
NIH-PA Author Manuscript
We reasoned that identifying genetic causes for the remaining 50% of DBA cases might
provide insight into the mysterious pathogenesis of this disorder. We recently reported the
first non-ribosomal gene mutated in DBA, which we identified by using whole-exome
sequencing.11 Specifically, we reported mutations in the GATA1 gene, which encodes a key
hematopoietic transcription factor essential for the specification of erythroid cells, as well as
megakaryocytes and eosinophils, from early hematopoietic stem and progenitor cells.11,12 In
humans, GATA1 mRNA is alternatively spliced to produce two forms of the protein – a long
(or full-length) form derived from inclusion of the second exon and a short form without this
exon that therefore lacks the N-terminal 83 amino acids.11,13 The GATA1 mutations that we
previously identified occur in the splice donor site of exon 2; they affect splicing by
impairing the production of the mRNA encoding the full-length form.
While intriguing, it remained unclear whether the GATA1 mutations act through a similar
mechanism as the haploinsufficiency for ribosomal proteins or represent a distinct subset of
DBA.7 In an attempt to address this question, we undertook a systematic screening for novel
GATA1 mutations in over 200 additional DBA patients and identified a highly informative
mutation in a male patient who had received a clinical diagnosis of DBA (Fig. 1 and
Supplementary Fig. 1, Supplementary Table 1). This distinct mutation changed the first
translation initiation codon in GATA1, ATG, to an ACG (Fig. 1a). Consistent with X-linked
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 3
NIH-PA Author Manuscript
inheritance, the patient’s asymptomatic mother was a carrier for this mutation (Fig. 1a).
When expressed exogenously in human 293T cells, wild-type GATA1 cDNA predominantly
produced the full-length form of the protein (Fig. 1c). In contrast, the ACG mutant cDNA
predominantly expressed the short form of GATA1 lacking the first 83 amino acids, but the
mutant cDNA also produced a low-level of full-length GATA1 (Fig. 1c, Supplementary
Figure 2), consistent with the observation that mammalian ribosomes are capable of
initiating translation at specific non-AUG triplets, including ACG.14 We ruled out the
possibility that the residual full-length GATA1 production resulted from contamination
through the use of independent clones and by sequencing (Supplementary Fig. 3). This
observation indicates that DBA can be caused by distinct mutations that reduce, but do not
entirely abolish, production of full-length GATA1. To gain further insight into the
physiological relevance of full-length GATA1 activity, we examined expression of GATA1
proteins during human erythropoiesis and observed that the full-length protein form appears
to be specifically upregulated in the course of erythroid differentiation (Fig. 1d).
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Our findings from an informative patient and studies of human hematopoiesis suggest that
the level of GATA1 full-length protein expression appears critical for the promotion of
normal erythropoiesis. This suggests the hypothesis that the more commonly observed
ribosomal protein mutations in DBA may result in anemia by reducing the protein
production of GATA1 and thus connect these two seemingly disparate sets of molecular
lesions. To test this hypothesis, we initially focused on the ribosomal protein gene, RPS19,
which is mutated in approximately 25% of DBA cases.1,2 We utilized a primary human
erythroid culture system to study haploinsufficiency of RPS19 using short hairpin RNAs
(shRNAs; Fig. 2a).15–17 We were able to reduce RPS19 protein expression and
concomitantly we noted reduced GATA1 protein expression (of both long and short forms)
with equal amounts of protein lysate loaded (Fig. 2a). While this primary cell system
primarily contains erythroid lineage cells, there is the possibility of differentiation to other
lineages and impairments in differentiation within the erythroid lineage, which could
therefore confound this analysis.16 We therefore turned to the clonal human erythroid K562
cell line to examine whether similar phenomena could be observed in a more homogenous
cell population. Upon knockdown of RPS19, we observed reduced GATA1 protein levels
(Fig. 2b) without significant effects on total cellular proteins, as assessed by Coomassie blue
staining of cell lysates (Fig. 2c). The levels of other erythroid-important proteins, including
TFRC, EPOR, JAK2, STAT5A, and TAL1, were also unchanged in the RPS19 knockdown
(Supplementary Fig. 4a). The reduction of both RPS19 and GATA1 protein levels could be
observed within 4 days of infection, before major effects on cell growth occurred
(Supplementary Fig. 4). Importantly, total GATA1 mRNA levels did not appear to be
affected (Fig. 2d), consistent with the decreased GATA1 protein levels being due to reduced
translation.
To directly assess whether the decrease in GATA1 protein levels was due to an effect on
translation, we performed polysome profiling and fractionation (Fig. 2e). This approach
allows for assessment of mRNA abundance in actively translating ribosomes that are present
in multiple copies on a single mRNA species, thereby forming polysomes indicative of
productive translation initiation.18,19 While the levels of GATA1 mRNA associated with
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 4
NIH-PA Author Manuscript
monosomes was similar or higher in RPS19 knockdown cells as compared with controls
(Supplementary Fig. 5), there was a 2–3-fold reduction in GATA1 mRNA abundance in
polysomes (Fig. 2f). In contrast, all erythroid-important mRNAs tested showed a different
pattern, with similar or increased abundance in larger polysomes following RPS19
knockdown (Fig. 2g). We also labeled cells with the methionine analog L-azidohomoalanine
as an alternative approach to demonstrate a selective reduction in GATA1 translation.20
When chemical detection of L-azidohomoalanine was performed on immunoprecipitated
GATA1, there was a reduction in the level of newly translated GATA1 in cells with reduced
RPS19 levels (Supplementary Fig. 6), while global translation was decreased to ~40% of
controls (Supplementary Fig. 7), consistent with the slowing of cell growth observed and the
known global reduction of translation (48–75%) in DBA patients.21,22
NIH-PA Author Manuscript
NIH-PA Author Manuscript
To gain insight into the mechanism by which ribosomal protein haploinsufficiency results in
the observed selective defect in translation, we examined the effect of reducing other DBAassociated ribosomal proteins. Upon reduction of RPL11, RPL5, or RPS24, which are
collectively mutated in ~15% of DBA cases,1 we consistently noted decreased protein levels
of GATA1 (Fig. 3a–c). The degree of reduction in GATA1 protein levels corresponded well
with the extent of knockdown observed for the various ribosomal proteins (Fig. 3a–c;
Supplementary Fig. 8). Highlighting the selectivity of this effect, other regulators of
erythropoiesis such as EPOR, TAL1, TFRC (CD71), and STAT5A showed no change with
these reductions, consistent with the results seen with RPS19 knockdown (Fig. 3a, b). The
similar defect in GATA1 translation resulting from reduction of diverse ribosomal proteins
from both 40 and 60 S subunits of the ribosome suggests that an impairment of translation
initiation may underlie this observation. This observation is consistent with the concept that
availability of free ribosomes is the rate-limiting step in translation initiation.23 Eukaryotic
translation initiation factors (eIFs) play a critical role in translation initiation. This process
begins as eIF4E initially interacts with the 5′ terminal cap of mRNAs and then binds eIF4G,
which in turn recruits multiple factors, including eIF4A and eIF3, and allows the ribosome
to overcome restrictive motifs in the 5′ untranslated region (5′ UTR) to scan for the
appropriate translation initiation AUG codon.24 We therefore used a selective inhibitor of
the eIF4E-eIF4G interaction (termed 4EGI-1) to examine whether this would also
selectively affect GATA1 translation.25–27 Strikingly, when K562 cells were treated with
4EGI-1 for 48 hours, a marked decrease in GATA1 protein was noted, while other proteins
remained largely unaffected (Fig. 3d). This shows that GATA1 mRNA has a more stringent
requirement for eIF-dependent translation initiation in comparison with a number of other
cellular transcripts. We have observed a decreased ratio of erythroid to non-erythroid (as
assessed using the phenotypic marker CD235a/glycophorin A) cells when RPS19, RPL11,
RPL5, or GATA1 are reduced in primary CD34+ cell cultures (Supplementary Figs. 9, 10),
which is a phenotype characteristic of DBA.15,16 When 4EGI-1 was added in increasing
concentrations to such cultures, a similar and dose-dependent decrease in the relative
amount of erythroid cells was observed (Fig. 3e and Supplementary Fig. 11). Importantly,
the observed phenotype occurred with a reduction in GATA1 protein levels, but without
changes in the levels of a number of ribosomal and other proteins (Fig. 3f). The decreased
erythropoiesis with 4EGI-1 treatment could be largely rescued by overexpression of GATA1
itself, suggesting that this is a critical target of this treatment (Supplementary Fig. 12).
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 5
NIH-PA Author Manuscript
Collectively, these results show that the reduction in GATA1 translation observed with
reduced ribosomal protein levels likely reflects a broad sensitivity of GATA1 mRNA to
impairments in translation initiation.
NIH-PA Author Manuscript
Cells can physiologically regulate gene expression through variation of the translation
initiation potential, since different mRNAs can have variable barriers to translation.23,24 A
number of studies have demonstrated that transcripts with highly structured 5′ UTRs are
more poorly translated and have a need for increased initiation potential by the
cell.23,24,28,29 As the 5′ end of human GATA1 mRNA has never been fully characterized, we
performed rapid amplification of the 5′ ends (5′-RACE) of human GATA1 transcripts from
both primary erythroid and K562 cells. This analysis revealed a previously unreported 5′ end
of GATA1 that was found on the majority of clones from both short and long isoform
mRNAs (Supplementary Fig. 13a, b). The 5′ end of human GATA1 mRNA is predicted to be
highly structured (Supplementary Fig. 13c) and thus translation may be more readily
impaired in settings where the translation initiation potential was reduced, such as with
ribosomal protein haploinsufficiency.23,29,30 In support of this, we found that the GATA1 5′
UTR restricted translation to a greater extent than other endogenous 5′ UTRs of similar
length using a reporter assay (Supplementary Fig. 13d). Additionally, other transcripts with
highly structured 5′ UTRs showed reduced association with larger polysomes
(Supplementary Fig. 14), while 5′ UTRs that were unstructured were found in the genes that
did not change or were upregulated in these polysomes (Fig. 2g). Therefore, downregulation
of GATA1 mRNA translation - one of a select group of transcripts with a highly structured 5′
UTR - may act as an Achilles’ heel during hematopoietic development to impair
erythropoiesis specifically in the context of DBA, consistent with our finding that rare
mutations in the GATA1 gene itself are capable of causing DBA. It is possible that other
undefined regulatory or structural motifs in the 5′ UTR of GATA1 mRNA may explain the
observed effect on translation in the setting of reduced ribosomal protein levels, but we have
been unable to identify any characteristic motifs in this sequence. Mouse Gata1 mRNA has
a much shorter and less structured 5′ UTR than that of humans,31 which may explain the
failure of ribosomal protein haploinsufficiency to downregulate translation of this gene and
cause major impairments of erythropoiesis in mouse models.8
NIH-PA Author Manuscript
To substantiate whether our findings are of relevance in vivo in DBA patients with
ribosomal protein gene mutations, we sought to assay GATA1 levels or activity. Given the
difficulty of obtaining a sufficient number of stage-matched erythroid cells from DBA
patients to examine GATA1 protein levels, we used RNA expression data to examine
whether there was an alteration among GATA1 transcriptional target genes. We reasoned
that if GATA1 protein levels were specifically reduced in these cells, the transcriptional
activity of this key hematopoietic transcription factor should be globally reduced. We
performed global gene expression profiling on sorted erythroid progenitors
(CD34+CD71highCD45RA− bone marrow mononuclear cells) from 3 DBA patients with
known RPS19 mutations and 6 matched controls. This surface phenotype is known to
selectively enrich for early erythroid progenitors (the erythroid burst-forming and colonyforming units, or BFU-Es and CFU-Es, respectively) and results in similar colony numbers
from healthy individuals and DBA patients.32,33 After normalization of the gene expression
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 6
NIH-PA Author Manuscript
NIH-PA Author Manuscript
data, the expression of both experimentally-derived and curated GATA1 up-regulated target
gene datasets was examined (Supplementary Fig. 15). The global gene expression profiles
showed a high degree of correlation between the DBA and control erythroid progenitors (R2
= 0.990; Fig. 4a), supporting the fact that comparable populations of cells were obtained
from the patients and controls. We then applied the gene set enrichment analysis (GSEA)
algorithm to assess whether the GATA1 target gene sets were either globally up or down
regulated in cells from patients.34 Consistent with our hypothesis, we noted significant
global down regulation of GATA1-target gene sets in the erythroid progenitors from DBA
patients compared with controls (Fig. 4c, d; Supplementary Fig. 15; Supplementary Tables
2–4). Of over 600 gene sets from other known transcription factor binding sites tested (see
Online Methods), none of these other sets demonstrated such a significant and global
downregulation, highlighting the selectivity of the effect observed for GATA1 activity. We
noted that targets from several other erythroid-important transcription factors and p53
showed no significant change, unlike what was observed for GATA1 targets
(Supplementary Fig. 16; Supplementary Table 5). Importantly, GATA1 mRNA levels were
comparable between DBA and control samples (Fig. 4b), providing support for the idea that
the altered GATA1 activity is mediated at the translational rather than transcriptional level.
These results demonstrate that in primary DBA patient samples with RPS19 mutations,
GATA1 activity is decreased – consistent with it underlying the erythroid differentiation
defect in this condition.
NIH-PA Author Manuscript
We then sought to examine whether the reduced GATA1 levels are sufficient to impair
erythropoiesis. When either ribosomal proteins or GATA1 are targeted with shRNAs, we
consistently observe increased apoptosis, decreased growth, and decreased erythroid
differentiation of primary hematopoietic cells (Supplementary Figs. 4, 9, 10). These findings
are entirely consistent with the observations made in DBA patients with mutations in the
ribosomal protein or GATA1 genes.11 We modified the GATA1 cDNA to contain an
optimized 5′ UTR that lacked known structural barriers and should be effectively
translated.24 When erythroid cells with reduced RPS19 levels were infected with lentiviruses
harboring this modified GATA1 cDNA, we were able to rescue apoptosis to levels close to
the baseline observed in these cells, while the mutant form of GATA1 that predominantly
produces the short form did not possess such robust activity in this rescue experiment
(Supplementary Fig. 17a, b). To assess whether GATA1 could rescue the defect in erythroid
differentiation observed in primary hematopoietic cells with reduced ribosomal protein
levels, we concomitantly transduced primary cultured CD34+ cells with shRNAs targeting
RPL11 or RPL5 and modified GATA1 cDNA or appropriate controls. We found that
introduction of GATA1 into these cells could improve erythropoiesis in the primary
hematopoietic cells (Supplementary Figs. 17c) in a comparable manner to the extent of
rescue observed with exogenous introduction of the deficient ribosomal protein gene itself,35
while the mutant version of GATA1 showed a reduced rescue of erythroid differentiation.
Importantly, we could show a 2–4-fold increase in the ratio of CD235a+ erythroid to
CD235a− non-erythroid cells when GATA1 lentivirus was introduced into primary bone
marrow mononuclear cells from DBA patients (Fig. 4e, f). Not only was the frequency of
erythroid cells increased, but there was an improvement in erythroid differentiation as
indicated by smaller cell size (reduced cell size characterizes more mature erythroid cells),
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 7
NIH-PA Author Manuscript
mature morphology, and increased expression of genes critical for terminal erythroid
maturation (Fig. 4g, h, i; Supplementary Fig. 18).36 These findings show that GATA1, while
not the only target of translation dysregulation in DBA, is an important factor in mediating
the erythroid-specific defect observed in this condition. Given the limitations in comparing
differentiation of DBA patient samples, which normally show delayed and impaired
maturation, with cells from normal individuals,37 the extent of this rescue is likely to be
partial and is difficult to quantitatively assess. Nonetheless, these results suggest that
modulation of GATA1 protein levels in the context of ribosomal protein haploinsufficiency
can improve erythropoiesis and this may lead to potential therapeutic approaches to
ameliorate DBA and related forms of anemia, such as that observed in 5q- myelodysplastic
syndrome.35
NIH-PA Author Manuscript
Our finding of impaired translation of GATA1 mRNAs by the ribosome in DBA adds to our
increasing knowledge of translational control in mediating cell-type specific gene
expression.5 We have shown that mutations in the GATA1 gene itself can cause DBA by
reducing full-length protein levels, while the more common ribosomal protein mutations in
this disease result in impaired erythropoiesis at least in part through the same mechanism of
decreased GATA1 protein production (Supplementary Fig. 19). While prior work has
implicated p53 as mediating some of the phenotypes in DBA and this pathway is regulated
by GATA1,7,15,38 we would note that the phenotypes that we observe in K562 cells occur in
the absence of p53,39 suggesting that cell-type specificity in DBA is likely mediated by
multiple pathways, including p53, that lie downstream of GATA1. Indeed, we observe that
p53 is upregulated upon reduction of RPS19 or GATA1 in primary hematopoietic cells,
suggesting that these factors act through a common pathway (Supplementary Fig. 20). More
generally, our findings suggest that 5′ UTRs of mRNAs with low affinity that are
inefficiently translated by the ribosome, would have further reduced translation under
conditions of limited ribosome abundance (or with impairments in translation initiation) and
thus would be selected against, such that synthesis of their encoded proteins would be more
selectively impaired.23,40 Altered translation has been implicated in diverse human diseases
including autism, cancer, and other blood disorders.26,27 Our finding of selectively impaired
translation resulting from ribosomal protein haploinsufficiency provides a paradigm for
understanding the previously mysterious cell-type specific defects observed in DBA and
conditions such as congenital asplenia,2,3 as well as childhood leukemia.4 In addition, the
deeper understanding of DBA pathogenesis that is presented here suggests potential
therapeutic avenues that would involve targeting of GATA1 protein production through
either gene therapy or small molecule approaches for amelioration of the anemia observed in
this disease.
NIH-PA Author Manuscript
ONLINE METHODS
Cell culture
293T cells were maintained in DMEM with 10% fetal bovine serum (FBS), 2 mM LGlutamine, and 1% penicillin/streptomycin (P/S). K562 cells were maintained at a density
between 0.1×106 and 1×106 cells per milliliter in RPMI 1640 medium supplemented with
10% FBS, 2 mM L-glutamine, and 1% P/S. Culture of primary human cells is described
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 8
below. Cells were incubated at 37°C with 5% CO2. Where indicated, cultures were
supplemented with 10–100 μM 4EGI-1 (sc-202597, Santa Cruz Biotechnology) or DMSO.
NIH-PA Author Manuscript
Culture of human adult peripheral blood-mobilized CD34+ progenitors was performed using
a two-stage culture method, as has been described previously.41,42 CD34+ cells were
obtained from magnetically sorted mononuclear samples of G-CSF-mobilized peripheral
blood from donors and were frozen after isolation. Cells were obtained from the Fred
Hutchinson Cancer Research Center, Seattle, USA. Cells were thawed and washed into PBS
with 1% FBS, pelleted, and then seeded in StemSpan SFEM medium (StemCell
Technologies, Inc.) with 1x CC100 cytokine mix (Stem Cell Technologies, Inc.) and 1%
P/S. Cells were maintained in this expansion medium at a density of 0.1×106 and 1×106
cells per milliliter, with medium changes every other day as necessary. Cells were kept in
expansion medium for a total of 5 days. After this expansion phase, the cells were reseeded
into StemSpan SFEM medium with 1% P/S, 20 ng/mL SCF (PeproTech, Inc.), 1 U/mL Epo
(Amgen), 5 ng/mL IL-3 (PeproTech, Inc.), 2 μM dexamethasone (Sigma-Aldrich), and 1
mM β-estradiol (Sigma-Aldrich). Cells were maintained in differentiation medium, with
medium changes every other or every third day as needed. Cells were maintained at a
density of 0.1×106 and 1×106 cells per milliliter.
NIH-PA Author Manuscript
Bone marrow mononuclear cells from Diamond-Blackfan anemia patients and healthy
controls were isolated from human bone marrow aspirates using Ficoll-Paque Plus
(17-1440-02, GE Healthcare) density gradient centrifugation. Mononuclear cells were
cultured in IMDM with 3% human AB plasma, 2% human AB serum, 1% P/S, 3 U/ml
Heparin, 200μg/ml Holo-Transferrin (Sigma-Aldrich), 10 μg/ml Insulin, 10ng/ml SCF
(PeproTech, Inc.), 1ng/ml IL-3 (PeproTech, Inc.), 2 μM dexamethasone (Sigma-Aldrich)
and 3 U/ml Epo (Amgen) similar to recently described protocols for culturing human
erythroid cells.43 Cells were cultured for 1–2 days before infection as described below and
analyzed 4–5 days after infection.
NIH-PA Author Manuscript
Bone marrow mononuclear cells from DBA patients were collected after appropriate
informed consent was obtained. Patient 1 is male and was 6 months old at the time of bone
marrow collection. The patient had been noted to have fussiness and pallor at 4 months of
age and a complete blood count performed at that time showed a hemoglobin level of 4.3
g/dL. The patient remained on monthly transfusions to avoid symptomatic anemia. The
patient had no physical anomalies. The erythrocyte adenosine deaminase level was elevated.
Patients 2 and 3 were both male and in their 40s at the time of bone marrow collection.
These patients both had a macrocytic anemia that was diagnosed in infancy. Both
individuals had required intermittent transfusions, but details of their medical history were
limited since only select records were available for review. Both Patient 2 and 3 were
steroid-dependent to prevent the need for transfusions. Neither individual had any physical
anomalies and erythrocyte adenosine deaminase levels were reportedly elevated. RPS19
mutations in these two individuals were c.3G>A Met1Ile and c.185G>A Arg62Gln,
respectively.
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 9
Lentiviral vectors and infection
NIH-PA Author Manuscript
The shRNA constructs targeting human RPS19 (sh913 and sh916, RefSeqID NM_001022),
human GATA1 (sh19-23, RefSeq ID NM_002049), human RPS24 (sh1-2, RefSeq ID
NM_001026), human RPL5 (sh1-5 RefSeq ID NM_000969) and human RPL11 (sh1-5
RefSeq ID NM_000975) were obtained from the Mission shRNA collection (SigmaAldrich). The constructs were in the pLKO.1-puro lentiviral vector. The sequences of the
shRNAs used in this study are listed in Supplementary Table 6.
As controls, the lentiviral vectors pLKO-GFP and pLKO.1s (the empty pLKO.1 vector with
a 1.2 kb stuffer element) were used. For rescue experiments, erythroid cells were cotransduced with shRNAs targeting ribosomal proteins with either the HMD control, HMDGATA1, or HMD-GATA1 mutant, which contain the respective cDNAs. Double transduced
cells were identified after puromycin selection and GFP expression driven by an IRES-GFP
from the HMD vector.
NIH-PA Author Manuscript
For production of lentiviruses, 293T cells were transfected with the appropriate viral
packaging and genomic vectors using FuGene 6 reagent (Promega) according to the
manufacturer’s protocol. The medium was changed the day after transfection to RPMI 1640
or StemSpan SFEM medium. After 24 h, this medium was collected and filtered at 0.45 μm
immediately prior to infection of primary hematopoietic or K562 cells. The cells were mixed
with viral supernatant in the presence of 8 μg/ml polybrene in a 6-well plate at a density of
250,000–500,000 cells per well. The cells were spun at 2000 rpm for 90 min at 22°C and left
in viral supernatant overnight. The medium was replaced the morning after infection.
Selection of infected cells was started 24 hours after infection with 1 μg/ml or 2 μg/ml
puromycin for primary hematopoietic cells and K562 cells, respectively. The infection
efficiency for pLKO-GFP infected cells was assessed by measuring the frequency of GFP+
cells by flow cytometry 48 hours post infection. Typically, the frequency of GFP+ cells for
primary hematopoietic cells and K562 cells were between 30–60% and >95%, respectively.
Quantitative RT–PCR
NIH-PA Author Manuscript
Isolation of RNA was performed using the miRNeasy Mini Kit (Qiagen). An on-column
DNase (Qiagen) digestion was performed according to the manufacturer’s instructions. RNA
was quantified by a NanoDrop spectrophotometer (Thermo Scientific). Reverse transcription
was carried out using the iScript cDNA synthesis kit (Bio-Rad). Real-time PCR was
performed using the ABI 7900 Machine Real-Time PCR system and SYBR Green PCR
Master Mix (Applied Biosystems). Quantification was performed using the ΔΔCt method.
Normalization was performed using β-actin mRNA as a standard, unless otherwise
indicated. The primers used for quantitative RT-PCR are listed in Supplementary Table 7.
5′RACE
5′ Rapid Amplification of cDNA ends (RACE) was conducted using total RNA from K562
cells and in vitro cultured primary erythroid cells using the FirstChoice RLM-RACE kit
(AM1700, Life Technologies) according to the manufacturer’s instructions. RACE PCR
products were subcloned using the TOPO TA cloning kit (450641, Invitrogen) followed by
Sanger sequencing. The primers used for 5′RACE are listed in Supplementary Table 7.
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 10
Western blotting
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Cells were harvested at indicated time points, washed twice in PBS, resuspended in RIPA
lysis buffer (50 mM Tris-HCl at pH 7.4, 150 mM NaCl, 0.1%SDS, 1%NP-40, 0.25%
sodium deoxycholate, 1 mM DTT) supplemented with 1x Complete Protease Inhibitor
Cocktail (Roche), and incubated for 30 min on ice. After centrifugation at 14,000 rpm for 10
min at 4°C to remove cellular debris, the remaining supernatant was transferred to a new
tube, supplemented with sample buffer, and incubated for 10 min at 70°C. Equal amounts of
proteins were separated by SDS gel electrophoresis using the NuPAGE Bis-Tris gel system
(Invitrogen) and MOPS running buffer. Subsequently, proteins were transferred onto a
PVDF membrane using NuPAGE transfer buffer (Invitrogen). Membranes were blocked
with 3% BSA-PBST for 1 h and probed with GATA1 goat polyclonal antibody (M-20,
sc-1234, Santa Cruz Biotechnology) at a 1:500 dilution, RPS19 mouse monoclonal antibody
(WW-4, sc-100836, Santa Cruz Biotechnology) at a 1:500 dilution, RPL5 goat polyclonal
(D-20, sc-103865, Santa Cruz Biotechnology) at a 1:500 dilution, RPL11 goat polyclonal
(N-17, sc-25931, Santa Cruz Biotechnology) at a 1:500 dilution, RPS20 goat polyclonal
(G-15, sc-55035, Santa Cruz Biotechnology) at a 1:500 dilution, RPS24 rabbit polyclonal
(ab102986, Abcam) at a 1:1000 dilution, EPOR rabbit polyclonal (M-20, sc-697, Santa Cruz
Biotechnology) at a 1:500 dilution, CD71 rabbit polyclonal (H-300, sc-9099, Santa Cruz
Biotechnology) at a 1:500 dilution, JAK2 rabbit polyclonal (Clone 06-255, Millipore) at a
1:750 dilution, STAT5A rabbit polyclonal (C-17, sc-835, Santa Cruz Biotechnology) at a
1:500 dilution, TAL1 goal polyclonal (C-21; sc-12984, Santa Cruz Biotechnology) at a
1:500 dilution, p53 rabbit monoclonal (7F5; 2527, Cell Signaling Technology) at a 1:1000
dilution, β-Actin mouse monoclonal (AC-15, Sigma) at a 1:2500 dilution, or GAPDH mouse
monoclonal antibody (6C5; sc-32233, Santa Cruz Biotechnology) at a 1:1000 dilution in 3%
BSA-PBST for 1 h at room temperature or overnight at 4°C. Membranes were washed four
times with PBST, incubated with donkey anti-mouse, anti-goat, or anti-rabbit peroxidasecoupled secondary antibodies (715-035-150, 705-035-147, and 711-035-152; Jackson
ImmunoResearch) at a 1:5000 to 1:10000 dilution in 3% BSA-PBST for 1 h at room
temperature, washed three times with PBST, and incubated for 1 min with Western
Lightning Plus-ECL substrate (Perkin Elmer). Proteins were visualized by exposure to
scientific imaging film (Kodak).
NIH-PA Author Manuscript
Flow cytometry analysis and apoptosis
For flow cytometry analysis, in vitro cultured erythroid cells were washed in PBS and
stained with propidium iodide (PI), 1:60 APC-conjugated Glycophorin A/CD235a (HIR2,
17-9987-42, eBioscience), 1:60 FITC-conjugated CD71 (OKT9, 11-0719-42, eBioscience),
1:60 PE-conjugated CD41a (HIP8, 12-0419-42, eBioscience) and 1:60 PE-conjugated
CD11b (ICRF44, 12-0118-42, eBioscience). For apoptosis analysis the Annexin V-APC
staining kit was used according to the manufacturer’s instructions (550474, BD
Pharmingen). FACS analysis was conducted on a BD Bioscience LSR II flow cytometer.
Data were analyzed using FlowJo 8.6.9 (TreeStar).
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 11
Flow cytometry activated cell sorting of human bone marrow populations
NIH-PA Author Manuscript
Mononuclear cells from human bone marrow aspirates were isolated by Ficoll-Paque Plus
(17-1440-02, GE Healthcare) density gradient centrifugation. Cells were preincubated with
human Fc receptor binding inhibitor (14-9161, eBioscience) and stained with propidium
iodide (PI) and antibodies against CD235a, CD41a and CD71 as described above. Cell
sorting was conducted on a BD Bioscience Aria I.
Cloning of UTR reporter constructs and luciferase reporter assay
NIH-PA Author Manuscript
For cloning of the luciferase reporter constructs the endogenous 5′ untranslated region of the
luciferase gene of the pGL3-Promoter vector (Promega) was replaced by the respective
5′UTRs of GAPDH, ACTB and GATA1. The 5′UTRs were inserted 14 bp downstream of
the major transcription initiation site of the SV40 promoter. 293T cells were seeded in a 24well plate at a density of 50,000 cells per well. For transfection of 293T cells, 350ng of the
respective luciferase reporter vectors were cotransfected with 35 ng per well of the pRLSV40 vector using the FuGene 6 reagent according to the manufacturer’s protocol
(Promega). Cells were incubated at 37°C with 5% CO2 until analysis at 48 h. For measuring
luciferase reporter activity, the Dual-Glo Luciferase assay system (E2920, Promega) was
used according to the manufacturer’s protocol. Briefly, cells were resuspended in Dual-Glo
Luciferase assay reagent and incubated at room temperature for 10–30 min until
measurement of firefly luminescence on a Safire 2 microplate reader (Tecan). Subsequently,
Dual-Glo Stop and Glo reagent was added to the suspension and incubated for 10–30 min at
room temperature until measurement of Renilla luciferase activity. For each sample, the
ratio of Firefly:Renilla luminescence was calculated and normalized to the signal of the
GAPDH-UTR construct.
Polysome profiling
NIH-PA Author Manuscript
K562 cells were incubated with 100 μg/ml of cycloheximide for 10 minutes at 37°C, washed
twice with ice-cold PBS containing 100 μg/ml of cycloheximide and lysed in 10 mM TrisHCl (pH 7.4), 5mM MgCl2, 100 mM KCL, 2 mM DTT, 100 μg/ml cycloheximide, 500
U/ml RNasin (Promega) and 1x Complete Protease Inhibitor, EDTA-free (Roche) by
passing the lysate through a 26 gauge needle 4 times. Polysomes were separated on a 10–
50% linear sucrose gradient containing 20 mM HEPES-KOH (pH 7.4), 5 mM MgCl2, 100
mM KCl, 2 mM DTT and 100 μg/ml cycloheximide and centrifuged at 36,000 rpm for 2
hours in a SW41 rotor in the XE-90 ultracentrifuge (Beckman Coulter). Gradients were
fractionated using a Bicomp Gradient Station fractionator with absorbance at 254 nm used to
visualize the gradients using the Econo UV monitor (BioRad). Fractions were collected and
phenol-chloroform extraction was performed to isolate RNA. β-Actin primers were used for
normalization of abundance from monosome and polysome gradient fractions, similar to
what has previously been described.44
Protein labeling and detection
Click-iT chemistry was conducted for metabolic labeling of proteins. Briefly, K562 cells
were washed with warm PBS and incubated in methionine-free RPMI media (R7513, Sigma
Aldrich) with 10% FBS and 2 mM L-Glutamine for 1 h at 37°C, 5% CO2 to deplete
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 12
NIH-PA Author Manuscript
methionine reserves. For labeling, Click-iT AHA (L-Azidohomoalanine, C10102, Life
Technologies) was added at a final concentration of 50 μM for 4 h at 37°C, 5% CO2. Cells
were harvested, washed twice in PBS, resuspended in RIPA lysis buffer (50 mM Tris-HCl at
pH 7.4, 150 mM NaCl, 0.1%SDS, 1%NP-40, 0.25% sodium deoxycholate) supplemented
with 1x Complete Protease Inhibitor Cocktail (Roche), and incubated for 30 min on ice.
After centrifugation at 14,000 rpm for 10 min at 4 °C to remove cellular debris, the
remaining supernatant was transferred to a new tube. 50 μg of protein lysate was used for the
Click reaction with tetramethylrhodamine alkyne (TAMRA, T10183, Life Technologies)
using the Click-iT protein reaction buffer kit (C10276, Invitrogen) in a total volume of 200
μl according to the manufacturer’s instructions. Total cell proteins were separated by SDSgel electrophoresis and TAMRA signal was detected on a Typhoon 9200 imager (Amersham
Biosciences) using 532 nm excitation and 580 nm longpass emission. After TAMRA
imaging, gels were post-stained with SYPRO Ruby Protein Gel Stain (S12001, Invitrogen)
and imaged on Typhoon 9200 imager using 473 nm excitation and 610 nm long-pass
emission.
Immunoprecipiation coupled to Click-iT on beads labeling
NIH-PA Author Manuscript
Whole cell lysate was used for immunoprecipitation conducted with a goat polyclonal
antibody against GATA1 (M-20, sc-1234, Santa Cruz Biotechnology) bound to Dynabeads
Protein G (10003D, Life Technologies) for 3 h with rotation at 4°C. The Dynabeadsantibody-antigen complex was washed three times with RIPA buffer, resuspended in 50 μl
RIPA buffer and the Click-iT reaction was performed using TAMRA alkyne and the ClickiT protein reaction buffer kit (C10276, Invitrogen) for 1 hour at 4 °C in a total reaction
volume of 200 μl. The immunoprecipitate was then washed once in RIPA buffer and then
bound proteins were eluted in 40μl 2xLDS buffer incubated at 70°C for 10min. Proteins
were separated by SDS-gel electrophoresis followed by TAMRA detection as described
above or western blot analysis using antibodies against GATA1.
Sanger sequencing and analysis
Sanger sequencing of patient samples was performed using standard PCR-based methods
and analyzed as has been described.45
NIH-PA Author Manuscript
Microarray gene expression analysis
Bone marrow mononuclear cell samples were obtained from three DBA patients with RPS19
mutations (including a codon 84 frameshift, Arg62Gln, and Leu131Arg mutations), as well
as 6 control samples from healthy individuals. CD71highCD34+CD45− populations from the
mononuclear cells were sorted as described46 and RNA was isolated from these samples.
The RNA was hybridized to Affymetrix HG-133A microarrays and the arrays were scanned
for further analysis. Subsequently, the array data was normalized using the RMA
normalization method as implemented in the Bioconductor package in R (http://
www.bioconductor.org/). This data is deposited under accession number GSE41817 in the
Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/).
GATA1 up-regulated target genes were derived from the GSE628 dataset using either the 21
hour time point vs. 0 hours (the majority of known GATA1 up-regulated targets have been
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 13
NIH-PA Author Manuscript
shown to be up-regulated at this time point) or 30 hours vs. 0 hours.47 Genes with log2
change of > 1.0 with a p-value of < 0.05 at the two sets of time points were used to derive
gene sets.42 In addition, GATA1 targets were derived from the TRANSFAC database
(http://www.gene-regulation.com/).48 These gene sets were used to run the gene set
enrichment analysis (GSEA) algorithm, as has been described.49 As controls, we utilized
gene sets that contain genes that share a transcription factor binding site defined in the
TRANSFAC (version 7.4, http://www.gene-regulation.com/) database48 using the C3
transcription factor target set from the MSigDB (http://www.broadinstitute.org/gsea/msigdb/
index.jsp).
Statistics
All pairwise comparisons were assessed using an unpaired two-tailed Student t-test, unless
otherwise discussed in the main text. Results were considered significant if the p-value was
< 0.05.
Study Approval
NIH-PA Author Manuscript
All patients or their families had provided written informed consent to participate in this
study. The institutional review boards at Boston Children’s Hospital and Massachusetts
Institute of Technology approved the study protocols.
Supplementary Material
Refer to Web version on PubMed Central for supplementary material.
Acknowledgments
We are grateful to the DBA patients and families for their inspiration and encouragement. We thank A.
Wakabayashi and J. Ulirsch for valuable assistance and D.G. Nathan, S.H. Orkin, D.A. Williams, S.T. Chou, G.W.
Bell, C.R. Walkley, J. Flygare, and D.P. Bartel for valuable comments and advice. We are grateful to L. Solomon,
L. Gaffney, and T. DiCesare for assistance with illustrations. This work was supported by the German National
Academic Foundation (to L.S.L.), NIH grants R01 HL107558 and K02 HL111156 and a grant from the DBA
Foundation (to H.T.G.), NIH grant P01 HL32262 (to H.F.L.), NIH grant U54 HG003067-09 (to E.S.L.), and NIH
grant R21 HL120791-01 and a March of Dimes Basil O’Connor Scholar Award (to V.G.S.).
References
NIH-PA Author Manuscript
1. Gazda HT, et al. Frameshift mutation in p53 regulator RPL26 is associated with multiple physical
abnormalities and a specific pre-ribosomal RNA processing defect in diamond-blackfan anemia.
Hum Mutat. 2012; 33:1037–1044. [PubMed: 22431104]
2. Lipton JM, Ellis SR. Diamond-Blackfan anemia: diagnosis, treatment, and molecular pathogenesis.
Hematol Oncol Clin North Am. 2009; 23:261–282. [PubMed: 19327583]
3. Bolze A, et al. Ribosomal protein SA haploinsufficiency in humans with isolated congenital
asplenia. Science. 2013; 340:976–978. [PubMed: 23579497]
4. De Keersmaecker K, et al. Exome sequencing identifies mutation in CNOT3 and ribosomal genes
RPL5 and RPL10 in T-cell acute lymphoblastic leukemia. Nat Genet. 2013; 45:186–190. [PubMed:
23263491]
5. Xue S, Barna M. Specialized ribosomes: a new frontier in gene regulation and organismal biology.
Nat Rev Mol Cell Biol. 2012; 13:355–369. [PubMed: 22617470]
6. Nathan DG, Clarke BJ, Hillman DG, Alter BP, Housman DE. Erythroid precursors in congenital
hypoplastic (Diamond-Blackfan) anemia. J Clin Invest. 1978; 61:489–498. [PubMed: 621285]
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 14
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript
7. Weiss MJ, Mason PJ, Bessler M. What’s in a name? J Clin Invest. 2012; 122:2346–2349. [PubMed:
22706300]
8. Horos R, et al. Ribosomal deficiencies in Diamond-Blackfan anemia impair translation of transcripts
essential for differentiation of murine and human erythroblasts. Blood. 2012; 119:262–272.
[PubMed: 22058113]
9. Torihara H, et al. Erythropoiesis failure due to RPS19 deficiency is independent of an activated
Tp53 response in a zebrafish model of Diamond-Blackfan anaemia. Br J Haematol. 2011; 152:648–
654. [PubMed: 21223253]
10. Jaako P, et al. Mice with ribosomal protein S19 deficiency develop bone marrow failure and
symptoms like patients with Diamond-Blackfan anemia. Blood. 2011; 118:6087–6096. [PubMed:
21989989]
11. Sankaran VG, et al. Exome sequencing identifies GATA1 mutations resulting in DiamondBlackfan anemia. J Clin Invest. 2012; 122:2439–2443. [PubMed: 22706301]
12. Orkin SH, Zon LI. Hematopoiesis: an evolving paradigm for stem cell biology. Cell. 2008;
132:631–644. [PubMed: 18295580]
13. Wechsler J, et al. Acquired mutations in GATA1 in the megakaryoblastic leukemia of Down
syndrome. Nat Genet. 2002; 32:148–152. [PubMed: 12172547]
14. Peabody DS. Translation initiation at an ACG triplet in mammalian cells. J Biol Chem. 1987;
262:11847–11851. [PubMed: 3040720]
15. Dutt S, et al. Haploinsufficiency for ribosomal protein genes causes selective activation of p53 in
human erythroid progenitor cells. Blood. 2011; 117:2567–2576. [PubMed: 21068437]
16. Flygare J, et al. Deficiency of ribosomal protein S19 in CD34+ cells generated by siRNA blocks
erythroid development and mimics defects seen in Diamond-Blackfan anemia. Blood. 2005;
105:4627–4634. [PubMed: 15626736]
17. Sankaran VG, et al. MicroRNA-15a and -16-1 act via MYB to elevate fetal hemoglobin expression
in human trisomy 13. Proc Natl Acad Sci U S A. 2011; 108:1519–1524. [PubMed: 21205891]
18. Rousseau D, Kaspar R, Rosenwald I, Gehrke L, Sonenberg N. Translation initiation of ornithine
decarboxylase and nucleocytoplasmic transport of cyclin D1 mRNA are increased in cells
overexpressing eukaryotic initiation factor 4E. Proc Natl Acad Sci U S A. 1996; 93:1065–1070.
[PubMed: 8577715]
19. Kondrashov N, et al. Ribosome-mediated specificity in Hox mRNA translation and vertebrate
tissue patterning. Cell. 2011; 145:383–397. [PubMed: 21529712]
20. Dieterich DC, Link AJ, Graumann J, Tirrell DA, Schuman EM. Selective identification of newly
synthesized proteins in mammalian cells using bioorthogonal noncanonical amino acid tagging
(BONCAT). Proc Natl Acad Sci U S A. 2006; 103:9482–9487. [PubMed: 16769897]
21. Cmejlova J, et al. Translational efficiency in patients with Diamond-Blackfan anemia.
Haematologica. 2006; 91:1456–1464. [PubMed: 17082006]
22. Garcon L, et al. Ribosomal and hematopoietic defects in induced pluripotent stem cells derived
from Diamond Blackfan anemia patients. Blood. 2013; 122:912–921. [PubMed: 23744582]
23. Shah P, Ding Y, Niemczyk M, Kudla G, Plotkin JB. Rate-limiting steps in yeast protein
translation. Cell. 2013; 153:1589–1601. [PubMed: 23791185]
24. Sonenberg, N.; Hershey, JWB.; Mathews, M. Translational control of gene expression. Cold
Spring Harbor Laboratory Press; Cold Spring Harbor, NY: 2000.
25. Moerke NJ, et al. Small-molecule inhibition of the interaction between the translation initiation
factors eIF4E and eIF4G. Cell. 2007; 128:257–267. [PubMed: 17254965]
26. Santini E, et al. Exaggerated translation causes synaptic and behavioural aberrations associated
with autism. Nature. 2013; 493:411–415. [PubMed: 23263185]
27. Gkogkas CG, et al. Autism-related deficits via dysregulated eIF4E-dependent translational control.
Nature. 2013; 493:371–377. [PubMed: 23172145]
28. Gebauer F, Hentze MW. Molecular mechanisms of translational control. Nat Rev Mol Cell Biol.
2004; 5:827–835. [PubMed: 15459663]
29. Babendure JR, Babendure JL, Ding JH, Tsien RY. Control of mammalian translation by mRNA
structure near caps. Rna. 2006; 12:851–861. [PubMed: 16540693]
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 15
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript
30. Kozak M. An analysis of vertebrate mRNA sequences: intimations of translational control. J Cell
Biol. 1991; 115:887–903. [PubMed: 1955461]
31. Kobayashi E, Shimizu R, Kikuchi Y, Takahashi S, Yamamoto M. Loss of the Gata1 gene IE exon
leads to variant transcript expression and the production of a GATA1 protein lacking the Nterminal domain. J Biol Chem. 2010; 285:773–783. [PubMed: 19854837]
32. Gazda HT, et al. Defective ribosomal protein gene expression alters transcription, translation,
apoptosis, and oncogenic pathways in Diamond-Blackfan anemia. Stem Cells. 2006; 24:2034–
2044. [PubMed: 16741228]
33. Lansdorp PM, Dragowska W. Long-term erythropoiesis from constant numbers of CD34+ cells in
serum-free cultures initiated with highly purified progenitor cells from human bone marrow. J Exp
Med. 1992; 175:1501–1509. [PubMed: 1375263]
34. Subramanian A, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting
genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005; 102:15545–15550. [PubMed:
16199517]
35. Ebert BL, et al. Identification of RPS14 as a 5q- syndrome gene by RNA interference screen.
Nature. 2008; 451:335–339. [PubMed: 18202658]
36. Hu J, et al. Isolation and functional characterization of human erythroblasts at distinct stages:
implications for understanding of normal and disordered erythropoiesis in vivo. Blood. 2013;
121:3246–3253. [PubMed: 23422750]
37. Moniz H, et al. Primary hematopoietic cells from DBA patients with mutations in RPL11 and
RPS19 genes exhibit distinct erythroid phenotype in vitro. Cell Death Dis. 2012; 3:e356.
[PubMed: 22833095]
38. Trainor CD, Mas C, Archambault P, Di Lello P, Omichinski JG. GATA-1 associates with and
inhibits p53. Blood. 2009; 114:165–173. [PubMed: 19411634]
39. Chylicki K, et al. p53-mediated differentiation of the erythroleukemia cell line K562. Cell Growth
Differ. 2000; 11:315–324. [PubMed: 10910098]
40. Lodish HF. Model for the regulation of mRNA translation applied to haemoglobin synthesis.
Nature. 1974; 251:385–388. [PubMed: 4421673]
41. Sankaran VG, et al. Human fetal hemoglobin expression is regulated by the developmental stagespecific repressor BCL11A. Science. 2008; 322:1839–1842. [PubMed: 19056937]
42. Sankaran VG, et al. MicroRNA-15a and -16-1 act via MYB to elevate fetal hemoglobin expression
in human trisomy 13. Proc Natl Acad Sci U S A. 2011; 108:1519–1524. [PubMed: 21205891]
43. Hu J, et al. Isolation and functional characterization of human erythroblasts at distinct stages:
implications for understanding of normal and disordered erythropoiesis in vivo. Blood. 2013;
121:3246–3253. [PubMed: 23422750]
44. Kondrashov N, et al. Ribosome-mediated specificity in Hox mRNA translation and vertebrate
tissue patterning. Cell. 2011; 145:383–397. [PubMed: 21529712]
45. Sankaran VG, et al. Exome sequencing identifies GATA1 mutations resulting in DiamondBlackfan anemia. J Clin Invest. 2012; 122:2439–2443. [PubMed: 22706301]
46. Lansdorp PM, Dragowska W. Long-term erythropoiesis from constant numbers of CD34+ cells in
serum-free cultures initiated with highly purified progenitor cells from human bone marrow. J Exp
Med. 1992; 175:1501–1509. [PubMed: 1375263]
47. Welch JJ, et al. Global regulation of erythroid gene expression by transcription factor GATA-1.
Blood. 2004; 104:3136–3147. [PubMed: 15297311]
48. Matys V, et al. TRANSFAC and its module TRANSCompel: transcriptional gene regulation in
eukaryotes. Nucleic Acids Res. 2006; 34:D108–110. [PubMed: 16381825]
49. Subramanian A, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting
genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005; 102:15545–15550. [PubMed:
16199517]
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 16
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Figure 1. GATA1 full-length protein production is necessary for human erythropoiesis and its
disruption results in DBA
(a) Sanger sequencing of the first initiator codon region from exon 2 of GATA1 in a male
DBA patient, the patient’s mother, and a healthy control.
(b) Bone marrow aspirate (above at 100X objective magnification with scale bar at bottom
right showing 10 μM distance) and biopsy (below at 50X objective magnification with scale
bar showing 30 μM distance) sections from marrow of the DBA patient.
(c) The production of full-length and short protein forms of GATA1 from the full-length
mRNA with exogenous expression of either no vector, HMD lentiviral vector, HMDGATA1 wildtype, or the HMD-GATA1 mutant vectors in 293T cells. Arrowheads indicate
the major protein forms of GATA1 full length and GATA1 short.
(d) Western blots against GATA1 (C-terminal antibody) with a β-actin loading control.
Arrowheads indicate the major protein forms of GATA1 full length and GATA1 short.
NIH-PA Author Manuscript
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 17
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Figure 2. Ribosomal protein haploinsufficiency results in reduced translation of GATA1
(a) Protein lysates from CD34+ derived erythroid progenitors at the CFU-E to
proerythroblast stage (day 4 of differentiation) from cells infected with RPS19 shRNA
vectors, sh916 and sh913. Arrowheads indicate the major protein forms of GATA1 full
length and GATA1 short.
(b) Protein lysates from K562 cells infected with RPS19 shRNA vectors, sh916 and sh913.
Arrowheads indicate the major protein forms of GATA1 full length and GATA1 short.
(c) Total protein lysates from control infected or RPS19 shRNA infected cells. Molecular
weight (MW) markers are shown on the left side of the gel.
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 18
NIH-PA Author Manuscript
(d) GATA1 mRNA levels were measured from RPS19 shRNA infected cells (normalized to
β-actin; here quantitative RT-PCR results from exons 5–6 are shown with similar results
obtained for other exons; n ≥ 3 per group).
(e) Polysome profiles from control or RPS19 shRNA treated cells 4 days following
infection. The 80S and polysomes are labeled. All graphs are separated on the arbitrary yaxis (derived from relative absorbance at 254 nm) for ease of visualizing the data with the xaxis showing distance along the sucrose gradient.
(f) Relative abundance of GATA1 mRNA in polysome fractions as assessed by RT-PCR
(normalized to β-actin; *** p < 0.0001). All data are shown as the mean ± the standard
deviation (n ≥ 3 per group).
(g) Relative mRNA abundance in larger polysomes (>4 ribosomes) compared to
monosomes. Data are normalized to β-actin. Data are plotted on a log10 scale for ease of
viewing the range of relative abundance of various erythroid-important mRNAs. The data
are shown as the mean ± the standard deviation (n ≥ 3 per group).
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 19
NIH-PA Author Manuscript
Figure 3. Defects in diverse ribosomal proteins and eukaryotic initiation factors impair GATA1
translation
NIH-PA Author Manuscript
(a, b, c) Western blots are shown from cells obtained 5 days following infection with
shRNAs targeting RPL11, RPL5, and RPS24. Arrowheads indicate the major protein forms
of GATA1 full length and GATA1 short.
(d) Protein levels after increased concentrations of eIF4E-eIF4G interaction inhibitor
4EGI-1 are used to treat K562 erythroid cells (48 hours of treatment). Arrowheads indicate
the major protein forms of GATA1 full length and GATA1 short.
(e) Erythropoiesis (as measured by CD235a+ cells) and myeloid cell production (as
measured by CD11b+ or CD41a+ cells) following treatment of primary hematopoietic cells
with 4EGI-1 (48 hours of treatment).
(f) Protein levels after increased concentrations of 4EGI-1 are used to treat CD34-derived
primary erythroid cells (48 hours of treatment). Arrowheads indicate the major protein forms
of GATA1 full length and GATA1 short.
NIH-PA Author Manuscript
Nat Med. Author manuscript; available in PMC 2015 January 01.
Ludwig et al.
Page 20
NIH-PA Author Manuscript
Figure 4. Global disruption of GATA1 transcriptional activity in DBA patients and rescue of
ribosomal protein haploinsufficiency with exogenous GATA1
NIH-PA Author Manuscript
NIH-PA Author Manuscript
(a) Mean gene expression values of DBA and control patient samples are plotted on a scatter
plot with a linear regression shown in red. The coefficient of determination is shown.
(b) Relative GATA1 mRNA levels are plotted from DBA and control samples.
(c, d) Enrichment profiles from GSEA comparing the relative expression of genes in DBA
patients versus controls and examining the distribution of GATA1 target genes in this data
(shown as black bars).
(e) Representative FACS plots on transduction of bone marrow mononuclear cells from a
DBA patient with GATA1 or HMD (empty) control lentiviruses.
(f) The ratio of erythroid (CD235a+) to non-erythroid (CD235a−) -cells is shown after
transduction of GATA1 for three independent DBA patients. The data are shown as the
mean ± the standard deviation (n = 3). (** p < 0.01; *** p < 0.001).
(g) Representative FACS forward scatter histogram plots (measuring cell size) of cultured
DBA patient cells transduced with control or GATA1 lentivirus. The mean forward scatter
intensity is shown ± the standard deviation (n = 3).
(h) Representative cytospin images of cultured DBA patient cells transduced with control or
GATA1 lentivirus. Transduced cells were sorted based on GFP expression and stained with
May-Grünwald-Giemsa staining. Scale bar = 10 μm.
(i) Gene expression analysis by RT-qPCR in DBA patient cells transduced with GATA1
relative to the empty vector control. The data are shown as the mean ± the standard
deviation (for 3 independent samples). (Normalized to β-actin; *** p < 0.001)
Nat Med. Author manuscript; available in PMC 2015 January 01.